Ida et al. Critical Care (2015) 19:107 DOI 10.1186/s13054-015-0825-9

RESEARCH Open Access Effects of terlipressin as early treatment for protection of brain in a model of haemorrhagic shock Keila Kazue Ida1,2*, Denise Aya Otsuki1, Adolfo Toshiro Cotarelli Sasaki1, Emilyn Silva Borges1, Letícia Urbano Cardoso Castro3, Talita Rojas Sanches3, Maria-Heloisa Massola Shimizu3, Lúcia Conceição Andrade3, José-Otávio Costa Auler Jr1, Alex Dyson4, Kenneth John Smith2, Joel Avancini Rocha Filho5 and Luiz-Marcelo Sá Malbouisson1

Abstract Introduction: We investigated whether treatment with terlipressin during recovery from hypotension due to haemorrhagic shock (HS) is effective in restoring cerebral perfusion pressure (CPP) and brain tissue markers of water balance, oxidative stress and apoptosis. Methods: In this randomised controlled study, animals undergoing HS (target mean arterial pressure (MAP) 40 mmHg for 30 minutes) were randomised to receive lactated Ringer’s solution (LR group; n =14; volume equal to three times thevolumebled),terlipressin(TERLIgroup;n =14; 2-mg bolus), no treatment (HAEMO group; n =12) or sham (n =6). CPP, systemic haemodynamics (thermodilution technique) and blood gas analyses were registered at baseline, shock and 5, 30, 60 (T60), 90 and 120 minutes after treatment (T120). After the animals were killed, brain tissue samples were obtained to measure markers of water balance (aquaporin-4 (AQP4)), Na+-K+-2Cl− co-transporter (NKCC1)), oxidative stress (thiobarbituric acid reactive substances (TBARS) and manganese superoxide dismutase (MnSOD)) and apoptotic damage (Bcl-x and Bax). Results: Despite the HS-induced decrease in cardiac output (CO) and hyperlactataemia, resuscitation with terlipressin recovered MAP and resulted in restoration of CPP and in cerebral protection expressed by normalisation of AQP4, NKCC1, TBARS and MnSOD expression and Bcl-x/Bax ratio at T60 and T120 compared with sham animals. In the LR group, CO and blood lactate levels were recovered, but the CPP and MAP were significantly decreased and TBARS levels and AQP4, NKCC1 and MnSOD expression and Bcl-x/Bax ratio were significantly increased at T60 and T120 compared with the sham group. Conclusions: During recovery from HS-induced hypotension, terlipressin was effective in normalising CPP and cerebral markers of water balance, oxidative damage and apoptosis. The role of this pressor agent on brain perfusion in HS requires further investigation.

* Correspondence: [email protected] 1Laboratório de Investigação Médica (LIM-08), Disciplina de Anestesiologia, Faculdade de Medicina, Universidade de São Paulo, Avenida Doutor Arnaldo, 455, 2° andar, sala 2120, Cerqueira César, São Paulo, SP 01246-903, Brazil 2Department of Neuroinflammation, Institute of Neurology, University College London (UCL), 1 Wakefield Street, 2nd floor, WC1N 1PJ, London, UK Full list of author information is available at the end of the article

© 2015 Ida et al.; licensee BioMed Central. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Ida et al. Critical Care (2015) 19:107 Page 2 of 14

Introduction rates in animal studies of haemorrhagic shock [14,21]. It Haemorrhagic shock is the leading cause of early death has been reported that terlipressin can improve CPP in in trauma patients [1]. During the pre-hospital period, patients with acute liver failure [17], septic shock [22] and haemorrhage contributes to death in 33% to 56% of traumatic brain injury with catecholamine-resistant cases, and it is the most common cause of death among shock [23]. However, the effects of terlipressin on those found dead upon the arrival of emergency medical cerebral haemodynamics during early treatment for services personnel [2]. Neurological signs such as altered haemorrhagic shock remain unclear, and no data are mental state, which typically includes obtundation, available comparing the effects of resuscitation with disorientation, confusion, agitation and irritability, cannot terlipressin with those of standard fluid. be neglected, because cerebral hypoperfusion is a conse- We hypothesised that early recovery of haemorrhagic quence in patients experiencing bleeding-associated shock-induced hypotension with terlipressin could restore hypotension [3-5]. In addition, animal studies of haemor- CPP and improve oxygenation of the brain. Therefore, the rhagic shock have shown that cerebral ischaemia with purpose of the present study was to investigate the effects cell damage begins at the onset of the haemodynamic of early administration of terlipressin on CPP and brain impairment [6-8]. tissue oxygen pressure (PbtO2), as well as on the regula- Under hypotensive conditions such as those in haemor- tion of tissue markers of water balance (that is, AQP4 and rhagic shock, cerebral perfusion pressure (CPP) is NKCC1), oxidative stress (that is, TBARS and MnSOD) sustained below the lower limits of autoregulation and apoptosis (that is, Bax and Bcl-x) within the brain in a [8], which is detrimental to brain tissue oxygenation porcine model of haemorrhagic shock. [5,7,9]. Cerebral ischaemia has been associated with dysregulation of aquaporin-4 (AQP4) and Na+-K+-2Cl − Materials and methods co-transporter (NKCC1) in the astrocytes [9,10] and Bcl-2 Ethical approval related apoptotic proteins in the neurons [11]. Oxidative This prospective randomised experimental study was stress is implicated in the neuronal apoptosis that occurs approved by the Comissão de Ética em Pesquisa of in haemorrhagic shock. It has been shown to accompany the Hospital das Clínicas at Faculdade de Medicina of increased lipid peroxidation within the brain, as reflected Universidade de São Paulo (067/11 and 280/13). by changes in the levels of thiobarbituric acid reactive substances (TBARS) and changes in the expression of Surgical preparation and monitoring antioxidant enzymes such as manganese superoxide Female Large White pigs (n =46) weighing 20 to 30 kg dismutase (MnSOD) [12]. were fasted for 12 hours with free access to water Standard resuscitation practice for haemorrhagic shock before the experiments. The animals were premedicated mandates use of high-volume crystalloids. However, such with ketamine (5 mg/kg intramuscular) and midazolam therapy can result in adverse effects such as interstitial (0.25 mg/kg intramuscular), and anaesthesia was induced oedema in the gut and cellular oedema in the heart [13], with propofol (7 mg/kg intravenous). After endotracheal increases in the inflammatory cytokine profile [14] and intubation, anaesthesia was maintained with isoflurane increased intracranial pressure (ICP) [8]. Crystalloids may vaporized in 40% oxygen, and the pigs were ventilated also fail to recover CPP and oxygenation within the brain (Fabius GS Premium; Dräger, Lübeck, Germany) with a [8,15]. Terlipressin is a synthetic, long-acting (4 to 6 hours) tidal volume of 8 ml/kg and positive end-expiratory pres- analogue of vasopressin. The structure of terlipressin con- sure of 5 cmH2O. The respiratory rate was adjusted to tains a peptide that represents the natural hormone lysine maintain normocapnia (partial pressure of carbon dioxide vasopressin, the innate vasopressin analogue in pigs. Its in arterial blood (PaCO2) between 35 and 45 mmHg). structure is very similar to human arginine vasopressin, Lactated Ringer’s solution (4 ml/kg/hr) and pancuronium but the synthetic drug is characterised by a more specific (0.3 mg/kg/hr) were administered continuously through- V1 agonistic effect (V1:V2 ratio =2.2:1) compared with out the experiments. Body temperature was maintained arginine-vasopressin (V1:V2 ratio =1:1). Terlipressin has at 38°C using a heated mat (Medi-Therm II; Gaymar been studied as a vasoactive drug in the management of Industries, Orchard Park, NY, USA). catecholamine-resistant arterial hypotension in septic Both femoral arteries were catheterised for measurement shock [16], liver failure [17] and acute gastrointestinal of mean arterial pressure (MAP) and withdrawal of blood bleeding [18]. The effects of terlipressin consist of to induce haemorrhagic shock, respectively. Another vasoconstrictive activity on vascular smooth muscle catheter was inserted into the right femoral vein for cells and a pronounced vasoconstriction within the later administration of treatments. splanchnic circulation that has been shown to redistribute A 7.5-French pulmonary artery catheter (Swan-Ganz; blood flow to recover perfusion pressure to organs such as Edwards Lifesciences, Irvine, CA, USA) was surgically the liver, kidney and brain [19,20] and to increase survival introduced into the right internal jugular vein and advanced Ida et al. Critical Care (2015) 19:107 Page 3 of 14

under continuous pressure recording into wedge position. Randomisation was previously performed, and the blind Cardiac output was determined by bolus pulmonary artery allocation of the pigs among groups was placed in thermodilution (Vigilance monitor; Edwards Lifesciences). numbered manila envelopes, which were opened in a All catheters and pressure transducers were filled with consecutive manner immediately before baseline mea- isotonic saline containing heparin (5 U/ml) and connected surements were registered. to a multiparametric data collection system (IntelliVue Haemorrhagic shock was induced by pressure-controlled MP50 monitor; Philips Healthcare, Best, the Netherlands). bleeding targeting a MAP of 40 mmHg, which was main- The heart rate (HR), right atrial pressure (RAP), mean tained for 30 minutes before treatment. Data were recorded pulmonary artery pressure (MPAP), pulmonary artery prior to blood removal (baseline), at 30 minutes after occlusion pressure (PAOP) and central body temperature achieving the target MAP (shock) and at 5 minutes (T5), were also continuously monitored with the IntelliVue 30 minutes (T30) and 60 minutes (T60) after treatment. In MP50 monitor. some of these animals (sham group: n =3; HAEMO group: The cardiac index (CI) was calculated to normalise the n =9; LR group: n =9; TERLI group: n =9), the study was data for body surface area in square metres by using a continued for 1 additional hour, allowing data to be conversion factor appropriate for pigs (k ×BW2/3,where registered at 90 minutes (T90) and 120 minutes k =0.09) [24]. Systemic vascular resistance index (T120) posttreatment. At the end of the study, the (SVRI), pulmonary vascular resistance index (PVRI), animals were killed with an overdose of isoflurane left ventricular stroke work index (LVSWI), right ven- and potassium chloride. The intraparenchymal probe tricular stroke work index (RVSWI), stroke volume then was macroscopically inspected for insertion index (SVI), systemic oxygen delivery index (DO2I), depth, and cortical samples of the brain were collected systemic oxygen consumption (VO2I) and systemic oxygen and immediately frozen in liquid nitrogen and stored at extraction ratio (O2ER) were calculated using standard −80°C for later analysis. formulae [25]. Arterial and mixed venous blood were sampled at Preparation of cerebral samples for Western blotting each time point for blood gas analysis, including assays and thiobarbituric acid reactive substance measurement of haemoglobin (Hb), lactate, sodium measurement (Na+) and potassium (K+) ion levels (ABL 555 blood gas The samples were homogenized in ice-cold solution meter; Radiometer, Copenhagen, Denmark). (200 mM mannitol, 80 mM 4-(2-hydroxyethyl)piperazine- Two burr holes of 5-mm diameter each were placed over 1-ethanesulfonic acid, 41 mM KOH, pH 7.5) containing a the right and left coronal sutures (12-mm paramedian). In protease inhibitor cocktail (Sigma-Aldrich, St Louis, the right hemisphere, an intraparenchymal probe was MO, USA) using a POLYTRON PT 10–35 homogenizer inserted into the cerebral cortex (15-mm depth) and (KINEMATICA, Lucerne, Switzerland). The homogenates secured with a single lumen bolt for measurement of were centrifuged at 4,000 × g for 30 minutes at 4°C to re- PbtO2 (Neurovent-PTO; RAUMEDIC, Helmbrechts, move cell debris. Protein concentrations were determined Germany). On the left side, a fibre-optic probe (Codman by the Bradford assay method using a Bio-Rad protein ICP EXPRESS Monitoring System; Codman Neuro, assay kit (Bio-Rad Laboratories, Hercules, CA, USA). Raynham, MA, USA) was inserted epidurally for con- tinuous monitoring of ICP after sealing the cranial Western blotting window with bone wax. CPP was calculated using a Western blotting assays were performed to assess the standard formula (CPP = MAP − ICP) [6]. expression of the following proteins: AQP4, NKCC1, Bcl-x, Bax and MnSOD. Cerebral tissue samples were Experimental design run on 12% polyacrylamide minigels for AQP4, Bax, Bcl-x Following surgical preparation, animals were allowed to and MnSOD and on 8% polyacrylamide minigels for stabilise for 30 minutes before being randomly divided NKCC1. After transfer by electroelution to polyvinylidene into one of the following four groups: (1) a sham group fluoride membranes (Amersham Hybond-P; GE Healthcare (n =6) consisting of animals that were not subjected to Life Sciences, Little Chalfont, UK), blots were blocked with haemorrhagic shock, (2) a HAEMO group (n =12) that 5% non-fat milk and 0.1% Tween 20 in Tris-buffered saline was subjected to haemorrhagic shock and did not receive for 1 hour. Blots were then incubated overnight with an treatment, (3) a LR group (n =14) that was subjected to anti-AQP4 antibody (1:2,000), NKCC1 (1:500), Bcl-x anti- haemorrhagic shock and treated with LR (volume equal to body (1:500), Bax antibody (1:500) and MnSOD antibody three times the volume bled) and (4) a TERLI group (1:200). The labelling was visualised with a horseradish (n =14) that was subjected to haemorrhagic shock and peroxidase-conjugated secondary antibody (anti-rabbit treated with terlipressin (2-mg bolus of GLYPRESSIN; immunoglobulin G (IgG), diluted 1:2,000; anti-goat IgG, Ferring Pharmaceuticals, São Paulo, Brazil). diluted 1:10,000; anti-mouse IgG diluted 1:2,000; or Ida et al. Critical Care (2015) 19:107 Page 4 of 14

anti-mouse IgG, diluted 1:2,000, respectively; Sigma- Aldrich) using an enhanced chemiluminescence (ECL) detection system (Amersham ECL Western Blotting Detection kit; GE Healthcare Life Sciences). The ECL membranes were scanned using Alliance 4.2 (UVItec, Cambridge, UK), and the cerebral AQP4, NKCC1, Bcl-x and Bax protein expression levels were quantified using densitometry, normalizing the bands to actin expression.

Thiobarbituric acid assay To assess the levels of TBARS, a 0.2-ml cortical cerebral homogenate sample was diluted in 0.8 ml of distilled water, followed by addition of 1 ml of 17.5% Figure 1 Percent survival at 120 minutes after haemorrhagic trichloroacetic acid. Then, 1 ml of 0.6% thiobarbituric shock using the Kaplan–Meier curve. Rats treated with lactated acid, pH 2, was added to the sample and placed in a Ringer’s solution (LR) and terlipressin (TERLI) after haemorrhagic boiling-water bath for 15 minutes. After the sample shock were compared with the non-treated rats (HAEMO) and a sham group that was not subjected to bleeding. *P =0.0007 indicates was allowed to cool, 1 ml of 70% trichloroacetic acid was significant difference compared with the sham group. added and the mixture was incubated for 20 minutes. The sample was then centrifuged at 2,000 × g for 15 minutes. The absorbance was recorded at 534 nm using a spectro- Haemodynamics photometer, and values were calculated by using a molar HR was significantly increased in the HAEMO, LR and extinction coefficient of 1.56 × 105 M/cm. The TBARS TERLI groups from the time of shock to T120 compared levels were then normalised to the total protein concen- with the sham group (P <0.001). From T30 to T120, HR tration, and the results are expressed as nanomoles per was significantly lower in the LR group than in the gram of protein. HAEMO group (P <0.05) (Figure 2). MAP was signifi- cantly decreased at shock in all study groups compared Statistical analysis with the sham group (P <0.001). In the HAEMO group, Physiological and neuromonitoring parameters and the MAP was significantly decreased from T5 to T120 arterial and mixed venous blood gas data were analysed compared with the other groups (P <0.001). Compared (GraphPad Prism version 5.03 for Windows; GraphPad with sham animals, MAP was significantly decreased at Software, La Jolla, CA, USA) across groups and time T60, T90 and T120 in the LR group (P <0.001) and using two-way analysis of variance (ANOVA) tests. at T5 in the TERLI group (P <0.001). No significant Tukey’s tests were used for post hoc analysis. Analyses of differences in this variable were observed from T30 survival were performed according to the Kaplan–Meier to T120 between the LR and TERLI groups (Figure 2). method and compared using Fisher’s exact test. The last The CI was significantly decreased from T5 to T120 observation carried forward imputation method was in the HAEMO and TERLI groups compared with applied throughout the study for the animals that the sham group (P <0.01). At the corresponding time died. Differences in the expression levels of AQP4, points, the CI was significantly higher in the TERLI NKCC1, Bcl-x, Bax, MnSOD and TBARS were ana- group than in the HAEMO group (P <0.05). In the lysed by one-way ANOVA followed by the Student- LR group, CI was higher than in the HAEMO and Newman-Keuls test, and the results are presented as TERLI groups (P <0.05) (Figure 2). mean ± standard error. For all analyses, P<0.05 was RAP was significantly decreased (P <0.001) and SVRI considered statistically significant. and PVRI were significantly increased (P <0.01) from shock to T120 in the HAEMO and TERLI groups Results compared with the sham group (P <0.01), whereas no The blood volume withdrawn from each group was similar, significant differences were observed in these variables averaging 60% of the estimated blood volume (HAEMO from T5 to T120 between the LR and sham groups. group: 1,083 ± 124 ml; LR group: 1,162 ± 203 ml; TERLI MPAP was significantly decreased in all groups from group: 1,011 ± 215 ml). In the HAEMO group, the number T5 to T120 compared with the sham group (P <0.05) of deaths following haemorrhage was significantly higher (Figure 2 and Table 1). In all study groups, PAOP at 120 minutes (six deaths at 41 ± 15 minutes after shock; was significantly decreased at shock compared with P =0.0007). At 120 minutes after shock, one animal from the sham group. In the LR group, PAOP was significantly the TERLI group had died (at 80 minutes after shock) increased at T5 and T120 compared with the HAEMO (Figure 1). group at these corresponding time points (Table 1). LVSWI, Ida et al. Critical Care (2015) 19:107 Page 5 of 14

Figure 2 Effects of terlipressin and lactated Ringer’s solution in a porcine model of haemorrhagic shock. LR, Rats treated with lactated Ringer’s solution after haemorrhagic shock; TERLI, Rats treated with terlipressin after haemorrhagic shock; HAEMO, Non-treated rats; Sham, Rats

not subjected to bleeding; PbtO2, Brain tissue oxygen tension; CPP, Cerebral perfusion pressure; ICP, Intracranial pressure; MAP, Mean arterial pressure; HR, Heart rate; CI, Cardiac index; SRVI, Systemic resistance vascular index; Hb, Arterial haemoglobin level; O2ER, Oxygen extraction ratio; SvO2,Mixed venous oxygen saturation; BE, Base excess. Data are expressed as mean ± standard error. *P <0.05 versus sham group. †P <0.05 versus HAEMO group. ‡P <0.05 versus LR group. Ida et al. Critical Care (2015) 19:107 Page 6 of 14

Table 1 Effects of haemorrhagic shock on haemodynamics at baseline, shock and at 5, 30, 60, 90 and 120 posttreatment Baseline Shock After treatment (min) 5 306090120 RAP (mmHg) Sham 8 ± 1 9 ± 2 8 ± 1 8 ± 1 9 ± 1 7 ± 1 6 ± 0 HAEMO 7 ± 2 2 ± 1* 2 ± 1* 2 ± 1* 2 ± 1* 2 ± 1* 2 ± 1* LR 7±1 2±1* 8±2† 7±2† 6±2*† 5±2† 5±2† TERLI 7 ± 2 2 ± 1* 2 ± 1*‡ 3±1*‡ 3±1*‡ 2±1*‡ 2±1*‡ MPAP (mmHg) Sham 19 ± 1 20 ± 1 20 ± 1 20 ± 1 20 ± 1 18 ± 1 18 ± 3 HAEMO 20 ± 3 16 ± 2* 17 ± 3* 17 ± 3* 17 ± 4* 16 ± 2* 16 ± 2* LR 19 ± 2 17 ± 2* 22 ± 5*† 20 ± 4† 19 ± 3† 19 ± 2† 19 ± 2† TERLI 20 ± 1 16 ± 2* 18 ± 3†‡ 18 ± 3† 18 ± 2† 19 ± 2† 19 ± 2† PAOP (mmHg) Sham 12 ± 1 11 ± 2 11 ± 2 11 ± 1 11 ± 2 9 ± 1 11 ± 2 HAEMO 11 ± 1 8 ± 2* 8 ± 2* 8 ± 1* 8 ± 1* 7 ± 1* 7 ± 1* LR 11 ± 1 9 ± 2* 11 ± 2† 10±2 9±2 9±2 10±2† TERLI 11 ± 1 8 ± 1* 8 ± 2 9 ± 1 9 ± 1 9 ± 1 9 ± 1 LVSWI (g-m/m2/beat) Sham 30 ± 7 35 ± 13 35 ± 10 35 ± 11 33 ± 6 32 ± 6 29 ± 5 HAEMO 34 ± 6 4 ± 1* 3 ± 1* 3 ± 1* 3 ± 1* 3 ± 1* 3 ± 1* LR 35 ± 4 4 ± 1* 19 ± 3*† 15 ± 4*† 14 ± 4*† 13 ± 4*† 12 ± 4*† TERLI 37 ± 7 4 ± 1* 8 ± 2*† 10 ± 3*† 10 ± 3*† 9±3*† 8±4*† RVSWI (g × m/m2/beat) Sham 6 ± 0 7 ± 1 6 ± 1 6 ± 1 5 ± 1 7 ± 6 ± 0 HAEMO 7 ± 1 2 ± 0* 2 ± 0* 1 ± 0* 1 ± 0* 1 ± 0* 1 ± 0* LR 7±1 2±0* 6±2*† 5±2*† 5±2*† 5±1*† 5±1*† TERLI 8 ± 1 2 ± 1* 3 ± 1*† 3±0*† 3±1*† 3±0*† 3±1*† SVI (ml/beat/m2) Sham 44 ± 11 40 ± 11 38 ± 11 39 ± 13 41 ± 8 41 ± 9 38 ± 11 HAEMO 42 ± 6 8 ± 2* 8 ± 2* 7 ± 1* 7 ± 1* 7 ± 1* 7 ± 1* LR 43 ± 4 8 ± 1* 28 ± 4*† 25 ± 5*† 24 ± 6*† 24 ± 6*† 23 ± 6*† TERLI 47 ± 7 9 ± 1* 12 ± 3* 13 ± 2*† 13 ± 2*† 13 ± 3*† 13 ± 3*† LR, Rats treated with lactated Ringer’s solution after haemorrhagic shock; TERLI, Rats treated with terlipressin after haemorrhagic shock; HAEMO, Non-treated rats; Sham, Rats not subjected to bleeding; RAP, Right atrial pressure; MPAP, Mean pulmonary artery pressure; PAOP, Pulmonary artery occlusion pressure; LVSWI, Left ventricular stroke work index; RVSWI, Right ventricular stroke work index; SVI, Stroke volume index. *P <0.05 versus sham group. †P <0.05 versus HAEMO group. ‡P <0.05 versus LR group.

RVSWI and SVI were significantly decreased in the LR and groups, DO2I was significantly increased from T5 to TERLI groups compared with the sham group (P <0.01) T120 and from T30 to T120, respectively, compared and significantly increased in the HAEMO group, with the HAEMO group (P <0.05). In the LR and compared with the sham group (P <0.01) (Table 1). TERLI groups, the VO2I was significantly increased from T5 to T120 compared with the sham group (P <0.05) and Blood gases, oxygenation and electrolytes significantly increased from T30 to T120 compared − pH, HCO3, BE, SvO2 and DO2I were significantly with the HAEMO group (P <0.05). The levels of Hb + + decreased, and O2ER, arterial lactate and K were and Na were significantly lower from T5 to T120 in the significantly increased, from shock to T120 in the LR group compared with the sham group (P <0.05). The HAEMO, LR and TERLI groups compared with the ratio of arterial oxygen partial pressure to fractional sham group (P <0.05) (Table 2). In the LR and TERLI inspired oxygen and the level of PaCO2 did not change Ida et al. Critical Care (2015) 19:107 Page 7 of 14

Table 2 Effects of haemorrhagic shock on blood gases, oxygenation and electrolytes at baseline, shock and at 5, 30, 60, 90 and 120 minutes posttreatment Baseline Shock After treatment (min) 5 306090120

PaCO2 (mmHg) Sham 47 ± 0 43 ± 3 41 ± 2 42 ± 1 42 ± 0 40 ± 0 41 ± 1 HAEMO 42 ± 3 39 ± 4 35.6 ± 7* 43 ± 8 38 ± 6 38 ± 8 38 ± 8 LR 42 ± 4 38 ± 4 42.3 ± 3 41 ± 3 40 ± 2 39 ± 2 39 ± 2 TERLI 43 ± 2 42 ± 4 42.9 ± 5† 47 ± 6† 48 ± 7† 45 ± 4† 44 ± 5 2 DO2I (ml/min/m ) Sham 540 ± 2 571 ± 107 501 ± 105 522 ± 108 553 ± 64 543 ± 95 518 ± 131 HAEMO 597 ± 81 243 ± 52* 212 ± 64* 190 ± 37* 192 ± 38* 194 ± 46* 190 ± 37* LR 580 ± 61 237 ± 73* 400 ± 49*† 375 ± 62*† 361 ± 66*† 344 ± 64*† 339 ± 59*† TERLI 581 ± 96 238 ± 27* 288 ± 19* 307 ± 43*† 313 ± 51*† 302 ± 55*† 299 ± 73*† 2 VO2I (ml/min/m ) Sham 121 ± 3 140 ± 17 126 ± 28 133 ± 13 139 ± 9 133 ± 21 143 ± 24 HAEMO 126 ± 27 154 ± 36 139 ± 40 119 ± 20 123 ± 39 126 ± 27 122 ± 21 LR 129 ± 33 152 ± 38 173 ± 37* 176 ± 24*† 182 ± 30*† 178 ± 24*† 187 ± 33*† TERLI 123 ± 19 152 ± 21 161 ± 8* 174 ± 22*† 176 ± 23*† 172 ± 23*† 170 ± 24*† K+ (mmol/L) Sham 4.4 ± 0.3 4.3 ± 0.0 4.0 ± 0.1 4.3 ± 0.1 4.4 ± 0.1 4.4 ± 0.5 4.5 ± 0.3 HAEMO 4.3 ± 0.4 5.6 ± 0.9* 5.2 ± 0.9 5.7 ± 0.9* 5.8 ± 0.8* 6.0 ± 0.6* 6.2 ± 0.5* LR 3.9 ± 0.2 5.8 ± 1.5* 4.0 ± 0.3† 4.4 ± 0.4† 4.9 ± 0.6*† 5.2 ± 0.7* 5.1 ± 1.1* TERLI 4.2 ± 0.2 5.3 ± 0.5* 5.4 ± 0.5* 5.0 ± 0.5 5.2 ± 0.5* 5.3 ± 0.6* 5.5 ± 0.7* Na+ (mmol/L) Sham 140 ± 1 140 ± 0 142 ± 1 140 ± 3 141 ± 1 139 ± 3 139 ± 1 HAEMO 139 ± 3 137 ± 2 137 ± 3 137 ± 2 137 ± 2 137 ± 2 136 ± 2 LR 141 ± 2 138 ± 4 137 ± 3 137 ± 3* 136 ± 2* 135 ± 3* 135 ± 2* TERLI 140 ± 3 139 ± 3 137 ± 2 138 ± 3 138 ± 3 139 ± 3 139 ± 5 LR, Rats treated with lactated Ringer’s solution after haemorrhagic shock; TERLI, Rats treated with terlipressin after haemorrhagic shock; HAEMO, Non-treated rats; Sham, Rats not subjected to bleeding; PaCO2, Partial pressure of dioxide carbon in the arterial blood; DO2I, Oxygen delivery index; VO2I, Oxygen consumption index; K+, Potassium ion levels; Na+, Sodium ion levels. *P <0.05 versus sham group. †P <0.05 versus HAEMO group. significantly in any group during the study (Figure 2 Aquaporin-4 and Na+-K+-2Cl− co-transporter and Table 2). At 60 minutes after shock, semiquantitative immunoblot analysis revealed a significant increase in the expression Neuromonitoring of AQP4 in the HAEMO group (179 ± 12% of sham, CPP, ICP and PbtO2 were significantly decreased from P =0.0086), which was not reversed by treatment with shock to T60 in the HAEMO group compared with the LR (196 ± 8% of sham, P =0.0047), but was fully restored sham animals (P <0.05). Both treatments with LR and by TERLI (125 ± 6% of sham). In the TERLI group, the TERLI were followed by a significant increase in CPP expression of AQP4 was significantly higher at 60 minutes compared with the HAEMO group (P <0.01), with CPP compared with the HAEMO and LR groups (P =0.0071). recovering to values not significantly different from those At 120 minutes, a significant upregulation of AQP4 of the sham group. The LR group had the largest increase was observed only in the LR group (217 ± 37% of in ICP, which was observed from T5 to T120 (P <0.05 sham, P =0.0084), which was significantly higher than in versus sham; P <0.001 versus HAEMO group; P <0.001 ver- TERLI group (117 ± 19% of sham, P =0.0169) (Figure 3). sus TERLI group). The TERLI group had no significant dif- No significant increase in the expression of NKCC1 ferences in ICP from T30 to T120 compared with the sham was observed in any group at 60 minutes after shock, group. Treatments with LR and TERLI recovered PbtO2 to but NKCC1 expression was significantly increased at values similar to those in the sham group (Figure 2). 120 minutes in the HAEMO group (237 ± 47% of sham, Ida et al. Critical Care (2015) 19:107 Page 8 of 14

Figure 3 Semiquantitative immunoblotting of membrane fractions prepared from cerebral samples. A densitometric analysis for the expression of aquaporin-4 (AQP4) and Na+-K+-2Cl− co-transporter (NKCC1) in samples from the sham, HAEMO, LR and TERLI groups is shown. Immunoblots reacted with anti-AQP4 revealed a 34 kDa band, and those reacted with anti-NKCC1 revealed a 135 to 170 kDa band. Data are expressed as mean ± standard error. *P <0.05 versus sham group. †P <0.05 versus HAEMO group. ‡P <0.05 versus LR group.

P =0.0234), which was fully restored by treatment with Bax and Bcl-x terlipressin (100 ± 1% of sham, P =0.0270) (Figure 3). The antiapoptotic Bcl-x protein was significantly upreg- ulated at 60 and 120 minutes after shock in the TERLI Manganese superoxide dismutase and thiobarbituric acid group (60 minutes: 197 ± 17% of sham, P =0.0038; reactive substances 120 minutes: 261 ± 48% of sham, P =0.0033), but not in The levels of TBARS were not significantly different the HAEMO group (60 minutes: 122 ± 6% of sham; from the sham group in any study group at 60 minutes 120 minutes: 32 ± 8% of sham) or LR group (60 minutes: after shock. However, at 120 minutes after shock, 92 ± 7% of sham; 120 minutes: 67 ± 18% of sham) these levels were clearly higher in the HAEMO group (Figure 5). The proapoptotic Bax protein was markedly (0.38 ± 0.05 nmol/mg of protein; P =0.0013) and LR upregulated at 60 and 120 minutes in the HAEMO group (0.31 ± 0.10 nmol/mg of protein; P =0.0167), but group (60 minutes: 347 ± 46% of sham, P =0.0088; not in the TERLI group (0.14 ± 0.01 nmol/mg of protein) 120 minutes: 190 ± 31% of sham, P =0.0154) and LR compared with the sham group (0.03 ± 0.01 nmol/mg of group (60 minutes: 339 ± 28% of sham, P =0.0021; protein). At 120 minutes after shock, the levels of TBARS 120 minutes: 154 ± 16% of sham, P =0.0129), signifi- in the TERLI group were significantly lower than in the cantly higher than the TERLI group (Figure 5). The HAEMO group (P <0.0001) and the LR group (P =0.0394) Bcl-x/Bax ratio was significantly decreased at 60 minutes (Figure 4). Animals treated with LR had the highest after shock in the HAEMO group (0.26 ± 0.03, P =0.0002) expression of MnSOD at 60 minutes after shock and LR group (0.20 ± 0.02, P <0.0001) compared with (245 ± 11% of sham, P <0.0001), whereas no significant the sham group (1.00 ± 0.05). At the corresponding changes in the expression of MnSOD were observed in the time, animals treated with TERLI had a Bcl-x/Bax other groups at the corresponding time point (HAEMO ratio (0.77 ± 0.19) significantly higher than that of the group: 157 ± 10% of sham; TERLI group: 125 ± 5% of HAEMO group (P =0.0393) and LR group (P =0.0235). sham). At 120 minutes after shock, the expression of At 120 minutes after shock, the Bcl-x/Bax ratio was MnSOD was significantly increased in the HAEMO group significantly higher in the TERLI group (2.40 ± 0.46) (237 ± 14% of sham, P =0.0081), which was not reversed than in the HAEMO group (0.17 ± 0.05, p =0.0098) by LR (244 ± 9% of sham, P =0.0009), but it was fully and LR group (0.40 ± 0.08, P =0.0054), but not with restored by TERLI (105 ± 16% of sham) (Figure 4). that in the sham group (1.00 ± 0.44) (Figure 5). Ida et al. Critical Care (2015) 19:107 Page 9 of 14

Figure 4 Changes in the level of thiobarbituric acid reactive substances and expression of manganese superoxide dismutase from cerebral samples from the sham, HAEMO, LR and TERLI groups. (a) Each bar represents the group mean ± standard error. (b) Immunoblots reacted with antibody against manganese superoxide dismutase revealing a 24 kDa band. *P <0.05 versus sham group. †P <0.05 versus HAEMO group. ‡P <0.05 versus LR group. HAEMO, No-treatment group; LR, Lactated Ringer’s solution group; TERLI, Terlipressin group.

Discussion signalling within the brain. Survival times were similar at Haemorrhagic shock can result in global cerebral 120 minutes after haemorrhagic shock among groups, with hypoxia caused by hypovolaemia and hypotension, the exception of the HAEMO group. However, whereas and the haemodynamic resuscitation must restore therapy with TERLI provided superior outcomes than LR CPP in order to prevent ischaemic injury within the with regard to measures of cerebral damage, it provided brain [6-8,26]. The results of the present study indicate inferior outcomes with regard to systemic or peripheral that early treatment with terlipressin can recover CPP measures of haemodynamics and tissue perfusion. after haemorrhagic shock and that the underlying As expected, CPP was not preserved at a blood pressure mechanisms include regulation of water and Na+ channels, below the cerebral autoregulation threshold [27], which inhibition of oxidative stress and decrease of apoptotic was followed by reduction in PbtO2 [8,28,29]. In the Ida et al. Critical Care (2015) 19:107 Page 10 of 14

Figure 5 Semiquantitative immunoblotting of membrane fractions prepared from cerebral samples. A densitometric analysis for the expression of Bax and Bcl-x in samples from the sham, HAEMO, LR and TERLI groups is shown. Immunoblots reacted with pro- and antiapoptotic Bax and Bcl-x proteins revealed 23 kDa and 26 kDa bands, respectively. Data are expressed as mean ± standard error. *P <0.05 versus sham group. †P <0.05 versus HAEMO group. ‡P <0.05 versus LR group. HAEMO, No-treatment group; LR, Lactated Ringer’s solution group; TERLI, Terlipressin group. non-treated animals, these derangements were associated increase in the expression of MnSOD [12,33,34]. Oxygen with upregulation of AQP4 and NKCC1. These proteins free radicals exert their pathophysiologic effects by directly play an important role in the formation of cellular oedema attacking lipids and proteins in the biologic membranes, in the brain by regulating water and Na+ transport which can cause cellular dysfunction and induce apoptotic through the sealing junctions of the blood–brain barrier cell death [35]. Indeed, in the present study, the exposure in response to cerebral ischaemia [10,30-32]. AQP4 acts of the brain to a high level of oxidative stress following by increasing water transport mainly in the pericapillary foot haemorrhagic shock was associated with a marked shift in process of astrocytes [9], and NKCC1 acts by increasing the Bcl-x/Bax ratio, indicating a loss of antiapoptotic abil- − secretion of Na+,Cl and water through an intact ity [36]. Bcl-x and Bax are proteins that play an important blood–brain barrier into the brain [10]. Therefore, it was role in determining the relative sensitivity of neuronal suggested that cerebral hypoperfusion was followed by subpopulations to ischaemia. Accordingly, previous ischaemic lesions in the HAEMO group. Cerebral hypo- studies showed that haemorrhagic shock can induce a perfusion might have contributed to an accumulation of significant oxidative stress in the brain [12,37] and that lipid peroxidation products, as reflected by the increased cerebral ischaemia can decrease the Bcl-x/Bax ratio [38]. levels of TBARS, which might induce a compensatory Furthermore, other markers of cerebral cellular damage Ida et al. Critical Care (2015) 19:107 Page 11 of 14

have also been described following haemorrhagic shock that was the case, then PbtO2 would not have recovered, in other studies, such as an increased level of glycerol in unless mitochondria were incapable of using oxygen, brain tissue [26] and increased plasma levels of S100B [7], allowing an increased availability of oxygen within the findings which support the presently reported results. tissue. As mitochondrial function was not assessed in The direct vasoconstrictive effect of terlipressin, reflected the present study, this explanation remains speculative. by increased MAP and SRVI, prevented an improvement in Also, oxidative damage would probably be associated CI but probably allowed for the redistribution of blood flow with inflammation, but this was not supported by a towards the cerebral circulation, leading to the restoration previous study in which researchers found an improved of CPP and PbtO2. Some effects cannot be differentiated inflammatory cytokine profile in rats treated with ter- from the common properties of any vasopressor, but it has lipressin compared with LR [14]. Nonetheless, the been suggested that terlipressin can induce a selective outcome was associated with an increase in the Bcl-x/Bax vasoconstrictor effect according to the distribution of ratio, suggesting that if any cerebral oxidative injury or V1 vasopressin receptors. This hypothesis is supported by ischaemia were present, it probably was not sufficiently previous studies in which researchers used norepinephrine severe to trigger antiapoptotic signalling [36,38] within the in models of haemorrhagic shock, which did not improve brain in terlipressin-treated animals. CPP or oxygenation [39,40]. Furthermore, another study Treatment with LR, however, was followed by a showed that vasopressin, the natural terlipressin analogue, discrepancy between the increments in PbtO2 that was resulted in a significantly higher increase of CPP com- not accompanied by recovery of CPP. This discrepancy can pared with norepinephrine [41]. As with the TERLI group, be attributed to the systemic third-spacing of crystalloids, an increased SRVI was also found in non-treated animals; which would also be consistent with the decline in MAP however, this was not accompanied by an increase in over the course of 120 minutes [14]. This explanation is MAP in the HAEMO group. The increase in MAP could supported by the overexpression of AQP4 in the LR group, have restored the cerebral autoregulation in animals which might indicate a compensatory mechanism to treated with terlipressin, which might explain the recovery eliminate excess water within the brain water [44]. in CPP and PbtO2. Another explanation for the recovery An increase in brain water content can also explain in CPP is the higher PaCO2 compared with the HAEMO the increase in ICP, which, in turn, is detrimental to group. Because PaCO2 has a linear positive correlation the restoration of CPP. This finding is also in line with cerebral blood flow [42], it could account for more with the decreased blood Na+ levels in the group cerebral vasodilation and hence better perfusion com- treated with LR compared with the sham group. pared with non-treated animals. However, it was unclear Another hypothesis is that a decrease in blood viscosity whether terlipressin also acted directly via V1 vasopressin after intravascular volume expansion, despite some differ- receptors within the brain [43], which might also explain ences in ICP, could increase cerebral blood flow [45] and the recovery in ICP and, in turn, CPP. Moreover, the ex- thus explain the similar PbtO2 values between LR and pression levels of both AQP4 and NKCC1 were also re- TERLI groups. The LR group had the largest increase in stored in animals treated with terlipressin, which also ICP, but whether it induced overexpression of the brain supports the finding that cerebral perfusion was recov- tissue markers of water balance, oxidative stress and ered. Despite the increase in SRVI, the fact that lactate, apoptosis is unknown. The fact is that regardless of O2ER and SvO2 were not significantly different between the mechanism(s) underlying the failure of full recovery of groups suggests that terlipressin did not impair periph- CPP in the LR group, it remains the case that the markers eral perfusion compared with the HAEMO group. In of oxidative stress, TBARS and MnSOD, were overex- fact, BE was less negative in the TERLI group. However, pressed in animals treated with LR. The fluid infusion could compared with the treatment with LR, terlipressin re- have carried the overproduction of reactive oxygen species sulted in inferior outcomes with regard to measures (ROS) throughout the tissue initiating a postischaemic of haemodynamics such as CI, RAP, SVRI and PVRI, reperfusion injury [36,46,47]. ROS regulate mechanisms via which might explain the death of one animal in the inflammatory pathways that ultimately can decrease vascu- TERLI group. lar resistance [35], allowing for an increase in brain volume As with the present data, an improvement in CPP has that could be partly responsible for unrecovered CPP in the also been described in patients with persistent arterial LR group. In fact, researchers in a previous study found an hypotension and acute liver failure [17], traumatic brain increased proinflammatory cytokine profile and severe injury [23] and septic shock [22] who were successfully hypotension (40 mmHg) in rats treated with LR after treated with terlipressin. In the present study, systemic haemorrhagic shock [14]. This oxidative stress could have hypoperfusion could have accounted for unreleased, and reduced the antiapoptotic trend of the Bcl-x/Bax ratio in thus undetected, products of oxidative stress within the the brain during postischaemic reperfusion [36,38]. These brain, which could have caused cell damage. However, if alterations in the expression of Bcl-x and Bax may indicate Ida et al. Critical Care (2015) 19:107 Page 12 of 14

that mitochondria were dysfunctional, as these proteins are  These cerebral improvements were observed for at part of the intrinsic mitochondria-related apoptotic least 2 hours after haemorrhagic shock in animals pathway. Therefore, similar to the considerations pre- treated with terlipressin. viously described for the terlipressin-treated animals, one hypothesis is that an increased PbtO could have Abbreviations 2 ANOVA: Analysis of variance; AQP4: Aquaporin-4; BE: Base excess; CI: Cardiac been caused by an increased availability of oxygen because index; CO: Cardiac output; CPP: Cerebral perfusion pressure; DO2I: Systemic dysfunctional mitochondria are not capable of using the oxygen delivery index; ECL: Enhanced chemiluminescence; HAEMO: oxygen available in the tissue. No-treatment group; Hb: Haemoglobin; HR: Heart rate; HS: Haemorrhagic shock; ICP: Intracranial pressure; IgG: Immunoglobulin G; LR: Lactated Ringer’s Some limitations of this study should be noted. First is solution group; LVSWI: Left ventricular stroke work index; MAP: Mean arterial the short observation time, which we used because the blood pressure; MnSOD: Manganese superoxide dismutase; MPAP: Mean + + − experiment was designed to determine the early cerebral pulmonary artery pressure; NKCC1: Na -K -2Cl co-transporter; O2ER: Oxygen extraction ratio; PaCO2: Partial pressure of carbon dioxide in the arterial effects observed during prehospital care, rather than to blood; PAOP: Pulmonary artery occlusion pressure; PbtO2: Brain tissue oxygen determine long-term functional neurologic outcome or tension; PVRI: Pulmonary vascular resistance index; RAP: Right atrial pressure; ROS: Reactive oxygen species; RVSWI: Right ventricular stroke work index; correlation to brain histopathology. Second, PbtO2 was SVI: Stroke volume index; SvO2: Mixed-venous oxygen saturation; measured locally with a probe placed in the cerebral cortex, SVRI: Systemic vascular resistance index; TBARS: Thiobarbituric acid reactive and therefore global ischaemia caused by a heterogeneous substance; TERLI: Terlipressin group; VO2I: Systemic oxygen consumption. distribution of PbtO2 may have been underestimated with Competing interests regard to the brain region analysed. Third, cerebral blood The authors declare that they have no competing interests. flow was not measured, because the purpose of the present study was to evaluate changes in CPP and oxygenation. Authors’ contributions Also, the anaesthetics used may be cerebroprotective and KKI and DAO conceived the study, designed the trial; obtained research funding; collected, analysed and interpreted the data; and drafted the mayhavedifferenteffectsinICPandMAP,butthesewere manuscript and contributed substantially to its revision. ATCS, ESB, LUCC, TRS minimised by having a sham group and another group not and MHH collected, analysed and interpreted the data and drafted the treated after haemorrhagic shock, which were both sub- manuscript and contributed substantially to its revision. LCA and LMSM conceived the study, designed the trial, obtained research funding, jected to the same anaesthetic protocol as the treated supervised the conduct of the trial and data collection, provided senior groups. In addition, some varieties caused by different vaso- advice on study design and statistical analysis, and drafted the manuscript pressin receptors in pigs (lysine vasopressin) and humans and contributed substantially to its revision. JOCA contributed substantially to analysis and interpretation of data, provided senior advice on study (arginine vasopressin) might have resulted in a different design and statistical analysis, and drafted the manuscript and contributed haemodynamic response to terlipressin. However, these substantially to its revision. AD and KJS analysed and interpreted data, differences do not interfere with the results of our study, provided senior advice on study design and contributed substantially to manuscript revision. JARF supervised the conduct of the trial and data because the study was aimed at investigating differences collection, contributed to data analysis and interpretation, provided senior between groups and changes over time rather than advice on study design and statistical analysis, and drafted the manuscript presenting absolute values. Finally, we used only terlipres- and contributed substantially to its revision. KKI takes responsibility for the article as a whole. All authors read and approved the final manuscript. sin as a vasopressor, and therefore we are unable to report whether different vasopressors would have yielded other Acknowledgements results. KKI, ATSC, LUCC and LMSM received grant support from Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP 11/14386-2, 11/23845-0, 12/04128-9 and 11/00348-1). KKI received grant support from Coordenação Conclusions de Aperfeiçoamento de Pessoal de Nível Superior (Capes PDSE-BEX Ciência Early treatment with terlipressin was effective at restoring sem Fronteiras 10646/12-4 and CAPES/PROAP 055/2012–Anestesiologia/FMUSP). CPP and preventing dysregulation of water balance, ESB received grant support from Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq 155528/2013-0 Pibic 2401). The authors and oxidative and apoptotic markers within the brain, are grateful for all the support provided by the laboratory technician Gilberto following haemorrhagic shock in our model. These de Mello Nascimento from LIM-8 and research assistants Claudia Alexandria results indicate that the role of this pressor agent on Pereira, Clayton dos Santos, Alexsandra Moraes and Eva Guiss from Anestesiologia FMUSP. brain perfusion in haemorrhagic shock requires further investigation. Author details 1Laboratório de Investigação Médica (LIM-08), Disciplina de Anestesiologia, Faculdade de Medicina, Universidade de São Paulo, Avenida Doutor Arnaldo, Key messages 455, 2° andar, sala 2120, Cerqueira César, São Paulo, SP 01246-903, Brazil. 2Department of Neuroinflammation, Institute of Neurology, University  Early treatment with terlipressin recovered cerebral College London (UCL), 1 Wakefield Street, 2nd floor, WC1N 1PJ, London, UK. 3Disciplina de Nefrologia, Faculdade de Medicina, Universidade de São Paulo perfusion pressure and brain tissue oxygen tension (LIM-12 HC-FMUSP), Avenida Doutor Arnaldo, 455, 3rd floor, Cerqueira César, after haemorrhagic shock in pigs. São Paulo, SP 01246-903, Brazil. 4Division of Medicine, University College 5  Terlipressin was effective for normalising cerebral London (UCL), Gower Street, WC1E 6BT, London, UK. Divisão de Anestesiologia, Hospital das Clínicas da Faculdade de Medicina da markers of water balance, oxidative damage and Universidade de São Paulo (HCFMUSP), Av. Dr. Enéas de Carvalho Aguiar,155, apoptosis after haemorrhagic shock. 8th floor, 05403-000, São Paulo, SP, Brazil. Ida et al. Critical Care (2015) 19:107 Page 13 of 14

Received: 2 November 2014 Accepted: 19 February 2015 23. Salluh JIF, Martins GAR, Santino MS, Araújo LV, Freitas GG, Verdeal JCR. Early use of terlipressin in catecholamine-resistant shock improves cerebral perfusion pressure in severe traumatic brain injury. Acta Anaesthesiol Scand. 2007;51:505–8. References 24. Holt JP, Rhode EA, Kines H. Ventricular volumes and body weight in 1. Bulger EM, May S, Kerby JD, Emerson S, Stiell IG, Schreiber MA, et al. mammals. Am J Physiol. 1968;215:704–15. Out-of-hospital hypertonic resuscitation after traumatic hypovolemic shock: 25. Drummheller OJ. Cardiopulmonary anatomy and physiology. In: Wyka KA, a randomized, placebo controlled trial. Ann Surg. 2011;253:431–41. Mathews PJ, Rutkowski J, editors. Foundations of respiratory care. 2nd ed. 2. Sauaia A, Moore FA, Moore EE, Moser KS, Brennan R, Read RA, et al. Clifton Park, NY: Delmar Cengage Learning; 2011. p. 116–70. Epidemiology of trauma deaths: a reassessment. J Trauma. 26. Meybohm P, Cavus E, Bein B, Steinfath M, Weber B, Scholz J, et al. 1995;38:185–93. Neurochemical monitoring using intracerebral microdialysis during systemic 3. Gutierrez G, Reines HD, Wulf-Gutierrez ME. Clinical review: hemorrhagic haemorrhage. Acta Neurochir (Wien). 2007;149:691–8. shock. Crit Care. 2004;8:373–81. 27. Hickey R, Albin MS, Bunegin L, Gelineau J. Autoregulation of spinal cord 4. Vincent JL, De Backer D. Circulatory shock. N Engl J Med. 2013;369:1726–34. blood-flow: is the cord a microcosm of the brain. Stroke. 1986;17:1183–9. 5. Taccone FS, De Backer D. Is cerebral microcirculation really preserved in 28. Chien JC, Jeng MJ, Soong WJ, Hwang B. Effects of fluid resuscitation on shock states? Crit Care Med. 2010;38:1008–9. cerebral tissue oxygenation changes in a piglet model of hemorrhagic 6. Meybohm P, Cavus E, Bein B, Steinfath M, Brand PA, Scholz J, et al. Cerebral shock. J Chin Med Assoc. 2011;74:448–54. metabolism assessed with microdialysis in uncontrolled hemorrhagic shock 29. Wan JJ, Cohen MJ, Rosenthal G, Haitsma IK, Morabito DJ, Derugin N, et al. after penetrating liver trauma. Anesth Analg. 2006;103:948–54. Refining resuscitation strategies using tissue oxygen and perfusion 7. Meybohm P, Cavus E, Dörges V, Weber B, Stadlbauer KH, Wenzel V, et al. monitoring in critical organ beds. J Trauma. 2009;66:353–7. Release of protein S100B in haemorrhagic shock: effects of small volume 30. Manley GT, Fujimura M, Ma TH, Noshita N, Filiz F, Bollen AW, et al. resuscitation combined with arginine vasopressin. Resuscitation. Aquaporin-4 deletion in mice reduces brain edema after acute water 2008;76:449–56. intoxication and ischemic stroke. Nat Med. 2000;6:159–63. 8. Cavus E, Meybohm P, Doerges V, Hugo HH, Steinfath M, Nordstroem J, et al. 31. Xiao F, Arnold TC, Zhang S, Brown C, Alexander JS, Carden DL, et al. Cerebral effects of three resuscitation protocols in uncontrolled Cerebral cortical aquaporin-4 expression in brain edema following cardiac haemorrhagic shock: a randomised controlled experimental study. arrest in rats. Acad Emerg Med. 2004;11:1001–7. Resuscitation. 2009;80:567–72. 32. Wallace BK, Jelks KA, O’Donnell ME. Ischemia-induced stimulation of cerebral 9. Papadopoulos MC, Verkman AS. Aquaporin-4 and brain edema. microvascular endothelial cell Na-K-Cl cotransport involves p38 and JNK Pediatr Nephrol. 2007;22:778–84. MAP kinases. Am J Physiol Cell Physiol. 2012;302:C505–17. 10. Yan YP, Dempsey RJ, Sun DD. Na+-K+-Cl− cotransporter in rat focal cerebral 33. Huang HF, Guo F, Cao YZ, Shi W, Xia Q. Neuroprotection by manganese ischemia. J Cereb Blood Flow Metab. 2001;21:711–21. superoxide dismutase (MnSOD) mimics: antioxidant effect and oxidative stress 11. Zhang TJHJ, Wen DX, Hang YN, Sieber FE. Hippocampus bcl-2 and bax regulation in acute experimental stroke. CNS Neurosci Ther. 2012;18:811–8. expression and neuronal apoptosis after moderate hypothermic 34. Chen Z, Chen HZ, Rhee P, Koustova E, Ayuste EC, Honma K, et al. Induction cardiopulmonary bypass in rats. Anaesth Analg. 2006;102:1018–25. of profound hypothermia modulates the immune/inflammatory response in 12. Guven H, Amanvermez R, Malazgirt Z, Kaya E, Doganay Z, Celik C, et al. a swine model of lethal hemorrhage. Resuscitation. 2005;66:209–16. Moderate hypothermia prevents brain stem oxidative stress injury after 35. Rushing GD, Britt LD. Reperfusion injury after hemorrhage: a collective hemorrhagic shock. J Trauma. 2002;53:66–72. review. Ann Surg. 2008;247:929–37. 13. Moon PF, Hollyfield-Gilbert MA, Myers TL, Kramer GC. Effects of isotonic 36. Tsai MC, Chen WJ, Ching CH, Chuang JI. Resuscitation with hydroxyethyl crystalloid resuscitation on fluid compartments in hemorrhaged rats. starch solution prevents nuclear factor κB activation and oxidative stress Shock. 1994;2:355–61. after hemorrhagic shock and resuscitation in rats. Shock. 2007;27:527–33. 14. Lee CC, Lee MT, Chang SS, Lee SH, Huang YC, Yo CH, et al. A comparison of 37. Chen G, You G, Wang Y, Lu M, Cheng W, Yang J, et al. Effects of synthetic vasopressin, terlipressin, and lactated ringers for resuscitation of colloids on oxidative stress and inflammatory response in hemorrhagic uncontrolled hemorrhagic shock in an animal model. PLoS One. shock: comparison of hydroxyethyl starch 130/0.4, hydroxyethyl starch 2014;9:e95821. 200/0.5, and succinylated gelatin. Crit Care. 2013;17:R141. 15. Urbano J, López-Herce J, Solana MJ, Del Castillo J, Botrán M, Bellón JM. 38. Krajewski S, Mai JK, Krajewska M, Sikorska M, Mossakowski MJ, Reed JC. Comparison of normal saline, hypertonic saline and hypertonic saline Upregulation of bax protein levels in neurons following cerebral ischemia. colloid resuscitation fluids in an infant animal model of hypovolemic shock. J Neurosci. 1995;15:6364–76. Resuscitation. 2012;83:1159–65. 39. Cavus E, Meybohm P, Dorges V, Stadlbauer KH, Wenzel V, Weiss H, et al. 16. Morelli A, Ertmer C, Rehberg S, Lange M, Orecchioni A, Cecchini V, et al. Regional and local brain oxygenation during hemorrhagic shock: a Continuous terlipressin versus vasopressin infusion in septic shock prospective experimental study on the effects of small-volume resuscitation (TERLIVAP): a randomized, controlled pilot study. Crit Care. 2009;13:R130. with norepinephrine. J Trauma. 2008;64:641–8. discussion 648–9. 17. Eefsen M, Dethloff T, Frederiksen HJ, Hauerberg J, Hansen BA, Larsen FS. 40. Meybohm P, Renner J, Boening A, Cavus E, Grasner JT, Grunewald M, et al. Comparison of terlipressin and noradrenalin on cerebral perfusion, Impact of norepinephrine and fluid on cerebral oxygenation in intracranial pressure and cerebral extracellular concentrations of lactate and experimental hemorrhagic shock. Pediatr Res. 2007;62:440–4. pyruvate in patients with acute liver failure in need of inotropic support. 41. Meybohm P, Cavus E, Bein B, Steinfath M, Weber B, Hamann C, et al. J Hepatol. 2007;47:381–6. Small volume resuscitation: a randomized controlled trial with either 18. Levacher S, Letoumelin P, Pateron D, Blaise M, Lapandry C, Pourriat JL. Early norepinephrine or vasopressin during severe hemorrhage. J Trauma. administration of terlipressin plus glyceryl trinitrate to control active upper 2007;62:640–6. gastrointestinal bleeding in cirrhotic patients. Lancet. 1995;346:865–8. 42. Grubb Jr RL, Raichle ME, Eichling JO, Ter-Pogossian MM. The effects of 19. Narahara Y, Kanazawa H, Taki Y, Kimura Y, Atsukawa M, Katakura T, et al. changes in PaCO2 on cerebral blood volume, blood flow, and vascular Effects of terlipressin on systemic, hepatic and renal hemodynamics in mean transit time. Stroke. 1974;5:630–9. patients with cirrhosis. J Gastroenterol Hepatol. 2009;24:1791–7. 43. Manning M, Stoev S, Chini B, Durroux T, Mouillac B, Guillon G. Peptide and 20. Stadlbauer KH, Wagner-Berger HG, Wenzel V, Voelckel WG, Krismer AC, non-peptide agonists and antagonists for the vasopressin and oxytocin Klima G, et al. Survival with full neurologic recovery after prolonged V1a,V1b,V2 and OT receptors: research tools and potential therapeutic cardiopulmonary resuscitation with a combination of vasopressin and agents. Prog Brain Res. 2008;170:473–512. epinephrine in pigs. Anesth Analg. 2003;96:1743–9. 44. Papadopoulos MC, Manley GT, Krishna S, Verkman AS. Aquaporin-4 21. Bayram B, Hocaoglu N, Atilla R, Kalkan S. Effects of terlipressin in a rat model facilitates reabsorption of excess fluid in vasogenic brain edema. FASEB J. of severe uncontrolled hemorrhage via liver injury. Am J Emerg Med. 2004;18:1291–3. 2012;30:1176–82. 45. Bouma GJ, Muizelaar JP. Relationship between cardiac output and cerebral 22. Bradley PG, Allen EK, Menon DK. Terlipressin for cerebral perfusion pressure blood flow in patients with intact and with impaired autoregulation. support in a patient with septic shock. Anaesthesia. 2004;59:619–20. J Neurosurg. 1990;73:368–74. Ida et al. Critical Care (2015) 19:107 Page 14 of 14

46. Exo JL, Shellington DK, Bayir H, Vagni VA, Janesco-Feldman K, Ma L, et al. Resuscitation of traumatic brain injury and hemorrhagic shock with polynitroxylated albumin, hextend, hypertonic saline, and lactated Ringer’s: effects on acute hemodynamics, survival, and neuronal death in mice. J Neurotrauma. 2009;26:2403–8. 47. Aksu U, Bezemer R, Yavuz B, Kandil A, Demirci C, Ince C. Balanced vs unbalanced crystalloid resuscitation in a near-fatal model of hemorrhagic shock and the effects on renal oxygenation, oxidative stress, and inflammation. Resuscitation. 2012;83:767–73.

Submit your next manuscript to BioMed Central and take full advantage of:

• Convenient online submission • Thorough peer review • No space constraints or color figure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution

Submit your manuscript at www.biomedcentral.com/submit